城市道路交通流运行状态评价应用研究  

Application Research on the Evaluation of Urban Road Traffic Flow State

作　　者：范馨月 黄大志 沈齐 FAN Xinyue;HUANG Dazhi;SHEN Qi(School of Mathematics and Statistics,Guizhou University,Guiyang,Guizhou 550025,China;Public Security Department of Guizhou Province,Guiyang,Guizhou 550000,China;Science and Technology Department,Guiyang Public Security Traffic Administration Bureau,Guiyang,Guizhou 550081,China)

机构地区：[1]贵州大学数学与统计学院,贵州贵阳550025 [2]贵州省公安厅,贵州贵阳550001 [3]贵阳市公安交通管理局科技处,贵州贵阳550081

出　　处：《数学建模及其应用》2020年第3期67-74,F0003,共9页Mathematical Modeling and Its Applications

基　　金：国家自然科学基金(11961008);贵州省科技计划项目(黔科合基础[2019]1122号);贵州大学线上线下混合式课程建设项目(XJG202060)。

摘　　要：针对当前城市路段交通流状态评价方法以及交通拥堵态势预测效果不佳的问题,对2018年“深圳杯”数学建模挑战赛D题解决方案中定义的路段消散指数进行改进,在城市路网中将其作为统一描述交通流状态的指标,自适应计算消散指数阈值评价交通流运行状态.并以路段为单位,基于动态贝叶斯网络预测路段的短时交通流状态,采用时空耗散BP神经网络模型对短时车流量进行预测.本文是2018年“深圳杯”数学建模挑战赛D题的后续研究成果,在实验城市贵阳市的中环路进行成果应用,对中环路的交通流运行状态进行评价监测.Aiming at the problem that the current evaluation method of urban link-based traffic flow status and the prediction effect of traffic congestion situation are not good,the link-based dissipation index defined in 2018"Shenzhen Cup"mathematical modeling challenge(question D:Research on road traffic flow state reconstruction based on multi-source detection data)is improved,which is used as the index to uniformly describe the traffic flow state of urban road network,and the index threshold is calculated adaptively to evaluate the state of traffic flow.Taking the road section as a unit,the short-term traffic flow state of the road section is predicted based on the dynamic Bayesian network,and the short time traffic flow is predicted by using the spatiotemporal dissipation BP neural network model.This paper is the follow-up research of the"Shenzhen Cup"mathematical modeling challenge in 2018.It is applied in the middle ring road of Guiyang,to evaluate and monitor the traffic flow status of the middle ring road in Guiyang.

关 键 词：道路交通 交通流运行状态 消散指数 动态贝叶斯网络 BP神经网络 

分 类 号：U491[交通运输工程—交通运输规划与管理]